Did I get a junk blue pill batch?

Post here first, or if you can't find a relevant section!
Post Reply
castortiu
Posts: 59
Joined: Tue Nov 07, 2017 8:34 am
Location: Seattle, WA

Did I get a junk blue pill batch?

Post by castortiu » Tue Dec 05, 2017 12:12 pm

Couple months ago I bought 1 blue pill from Amazon to start working on STM while I ordered 10 more from AliExpress.

Yesterday started to include RTCClock into my project and the blue pill where I develop (1st one form Amazon) freeze all the time when calling getTime(), didn't know why, then tried the example Test_RTClock.ini and same thing, then I soldered the pin headers on another new blue pill (AliExpress), I tried and this time getTime() always returns 0 in every call, tried a new one and same behavior (0 every time from getTime()), then I tried on the 4th blue pill and the Test_RTClock.ino worked as expected and I was getting the time every second.

I still have 7 blue pills that haven't tried, however the RTC only works 1 in 4 so far.

Did I get a bad batch of blue pills or in general blue pills are kind of junk? for $2 dollars I was not expecting a lot but didn't expect so high failure rate.

ag123
Posts: 854
Joined: Thu Jul 21, 2016 4:24 pm

Re: Did I get a junk blue pill batch?

Post by ag123 » Tue Dec 05, 2017 12:15 pm

You probably aren't alone
viewtopic.php?f=3&t=2602

And oh there were some rather inquisitive discussion about that crystal, among which it may not oscillate
viewtopic.php?t=594
viewtopic.php?t=671

stevestrong
Posts: 2056
Joined: Mon Oct 19, 2015 12:06 am
Location: Munich, Germany
Contact:

Re: Did I get a junk blue pill batch?

Post by stevestrong » Tue Dec 05, 2017 1:54 pm

The best test to see if the blue pill works is to flash the new bootloader with STLink or serial in DFU mode.
If all this pass, and after reset you see the short LED blinks, and the PC recognizes it as USB serial COM port, then you are lucky.

User avatar
mrburnette
Posts: 2197
Joined: Mon Apr 27, 2015 12:50 pm
Location: Greater Atlanta
Contact:

Re: Did I get a junk blue pill batch?

Post by mrburnette » Tue Dec 05, 2017 3:41 pm

I am not surprised.

Out of 20 Maple Mini Baite clones, 19 worked without issue and 1 required hot-air reflow.

But "work" is not a really complete picture: digital I/O almost always work but analog inputs have been problematic. First, these cheap boards are 2-sided and do not have a separate analog and separate digital ground plane layers. Often, a thorough cleaning with standard 100% IPA will do the trick (followed by a baking in my home oven (when the wife is away.) There are some approved aerosols that probably will work better than IPA but since fluorocarbons were banned, nothing works as well as the old cleaners.

The 32K crystal quality has already been mentioned, but not all "watch crystals" are created equal, so when you discover one that works consistently, do a repeat buy. PDF Warning: NIST Electrical Characteristics of Quartz-Crystal

Ray

castortiu
Posts: 59
Joined: Tue Nov 07, 2017 8:34 am
Location: Seattle, WA

Re: Did I get a junk blue pill batch?

Post by castortiu » Tue Dec 05, 2017 5:19 pm

What is DFU mode?

That is a good idea, I'll give them a good clean and reflow on the toster oven I have for SMDs, what is the worst it can happen :D

castortiu
Posts: 59
Joined: Tue Nov 07, 2017 8:34 am
Location: Seattle, WA

Re: Did I get a junk blue pill batch?

Post by castortiu » Tue Dec 26, 2017 7:59 am

Found the issue with the RTC clock on my blue pills.

8 of the 10 blue pills came with the Crystal reverse soldered, took me a while to find out, since I never would have thought they would have solder a component in wrong direction. :roll:

Looks like the quality control for the blue pill is Zero.

Later I found the crystal datasheet and confirmed that the crystal is wrong, the crystal package is deceiving since usually the dot on top means the #1 pin, however in this case the 45 degrees edge is the pin 1 and the dot is in the pin #2,#3

http://www.datasheetarchive.com/pdf/dow ... +32.768kHz

I have the 10 blue pills RTC working now with the external oscillator RTCSEL_LSE.
Attachments
Before.jpg
Before.jpg (246.35 KiB) Viewed 252 times
Solder Mask.jpg
Solder Mask.jpg (204.55 KiB) Viewed 252 times
After.jpg
After.jpg (227.94 KiB) Viewed 252 times

User avatar
ahull
Posts: 1721
Joined: Mon Apr 27, 2015 11:04 pm
Location: Sunny Scotland
Contact:

Re: Did I get a junk blue pill batch?

Post by ahull » Tue Dec 26, 2017 9:31 am

Well spotted. I'm surprised the manufacturer didn't make these devices with no particular orientation since they only have two active pins do they not?
It seems that for some reason they put a two pin device in a four pin can... Example => http://www.chinafronter.com/wp-content/ ... %974.6.pdf
Very odd.
- Andy Hull -

castortiu
Posts: 59
Joined: Tue Nov 07, 2017 8:34 am
Location: Seattle, WA

Re: Did I get a junk blue pill batch?

Post by castortiu » Tue Dec 26, 2017 8:12 pm

Yes, from the datasheet pin #2 and #3 are not connected, and crystals are not polarized so they easily could have done bullet proof orientation, however I wonder if it depends of the crystal brand they get, may be in some brands pin #2 or #3 have some functionality and they need these pins not connected.

These boards are built by pick-and-place machines and usually the components are in reels, so they never get the wrong orientation, looks like they fed the machine with some reels left overs and the guy put it backwards.

About the package I wonder the same thing why they use a 4-pin package for a 2-pin crystal.

User avatar
ahull
Posts: 1721
Joined: Mon Apr 27, 2015 11:04 pm
Location: Sunny Scotland
Contact:

Re: Did I get a junk blue pill batch?

Post by ahull » Wed Dec 27, 2017 10:11 am

Perhaps some variants of this package include the load capacitors. One of the suppliers I looked at states "Package: Ceramic resonators come with standard, single-in-line package (ZTA, ZTB and ZTT), or 3-leads with built-in capacitors (ZTTS, ZTTCV)." , so I suspect this particular device comes in different flavours.
- Andy Hull -

pokemon99
Posts: 50
Joined: Wed Jun 07, 2017 9:09 am

Re: Did I get a junk blue pill batch?

Post by pokemon99 » Thu Dec 28, 2017 12:18 pm

I also face the problem work of the clock.
The first thing you need to wash off the flux is very good.
Under the crystal there can be a lot of flux that is not washed, it is necessary to blow out.
Oscilloscope is not present, therefore for check of working capacity of a crystal, I do so. When I take the crystal I take the ds3231 clock, it has an output of 32k, I alternately connect it to PC15 and PC14 if at one of their outputs the MCU does not start laundering yet until it starts on both outputs. After that, solder the crystal if there is no start, the crystal is replaced.

Post Reply